Autoimmune hepatitis experimental model based on adenoviral infections


  • Potential conflict of interest: Nothing to report.

To the Editor:

We read with great interest the article by Hardtke-Wolenski et al.[1] describing the development of an animal model of autoimmune hepatitis based on a self-limited adenoviral infection. The adenovirus administered encoded for formiminotransferase cyclodeaminase (FTCD), a targeted liver antigen in type 2 autoimmune hepatitis (AIH), identified in 1999. This report confirms our previously published findings that a self-limited adenoviral infection, with a virus encoding for FTCD, can lead to the development of an AIH in mice.[2]

The researchers state that danger signals” are necessary for the initiation of an autoimmune response against the liver based on adenoviral infections and hydrodynamic transfection experiments with an observation period of 12 weeks. These results are in contrast with previous findings in models of AIH generated by DNA vaccination[3] or adoptive transfer,[4] where a peripheral activation of T-cell specific to a liver autoantigen, in the absence of inflammation (danger signals), led to an active autoimmune response.

The researchers describe the development of fibrosis in their model, but it is solely based on silver staining of liver sections. It should be remembered that silver staining of reticulin proteins mainly reflects changes in the liver structure (as in Fig. 2C), where mild alterations are observed. These can be interpreted as the result of hepatocyte lysis secondary to the lymphocyte infiltration. Trichrome staining would have allowed one to visualize collagen deposition, the hallmark of liver fibrosis.

The researchers bring up an interesting point when they discuss the need for a predisposing genetic background (nonobese diabetes, in this case) for the development of an AIH in mice, an observation we previously reported on in our model of type 2 AIH. However, the complete absence of an AIH in C57BL/6 and FVB/N mice in their model is rather puzzling. We[2] and others[5] found that AIH can be triggered in both these mouse strains. This could be attributed to the duration of the observation period, which is critical in view of the fact that we observed the development of AIH as late as 8 months after adenoviral infection.[2] In the article, it is not clear whether the C57BL/6 and FVB/N mice were followed for more than 12 weeks. It could also be the result of the plaque-forming units of adenovirus used, an important factor in the outcome of an adenoviral infection or their adenovirus construction, because our vector encoded for CYP2D6 in addition to FTCD,[2] whereas Holdener et al. used an adenoviral vector encoding for CYP2D6.[5]

  • Pascal Lapierre, Ph.D.1

  • Kathie Béland, M.Sc.2

  • Fernando Alvarez, M.D.2

  • 1Department of Immunovirology, INRS-Institut Armand-Frappier, Laval, Quebec, Canada

  • 2Department of Gastroenterology, Hepatology, and Nutrition, CHU Sainte-Justine, Montréal, Quebec, Canada